J Antimicrob Chemother 2016; 71: 58 – 62
doi:10.1093/jac/dkv313 Advance Access publication 21 October 2015
Transmission of MRSA between humans and animals on duck
and turkey farms
E. van Duijkeren*, P. Hengeveld, T. P. Zomer, F. Landman, T. Bosch, A. Haenen and A. van de Giessen
Centre for Infectious Disease Control (CIb), National Institute for Public Health and the Environment (RIVM), PO Box 1,
3720 BA Bilthoven, The Netherlands
*Corresponding author. Tel: +31-(0)30-2743942; Fax: +31-(0)30-2744434; E-mail: [email protected]
Received 2 June 2015; returned 14 July 2015; revised 19 August 2015; accepted 31 August 2015
Objectives: The objectives of this study were to estimate the prevalence of MRSA on duck and turkey farms,
identify risk factors for human carriage and study transmission between animals and humans.
Methods: On 10 duck and 10 turkey farms, samples were taken from animals, poultry houses, home residences
and humans and cultured using pre-enrichment and selective enrichment. MRSA isolates were typed by spa typing and multiple-locus variable number tandem repeat analysis (MLVA) typing. A subset of isolates from animals
and humans was investigated using whole-genome mapping.
Results: MRSA was found on one duck farm and three turkey farms. On duck farms, all humans were MRSA negative. On turkey farms, 5 of 11 farmers, 2 of 32 family members and 15 of 49 samples from the home residences
were MRSA positive. Individuals with daily contact with turkeys were significantly more often MRSA positive than
individuals without daily contact. All MRSA isolates belonged to livestock-associated MLVA complex 398,
belonged to spa type t011, were negative for Panton – Valentine leucocidin, were mecC negative and were
mecA positive. Whole-genome mapping proved a valuable tool to study the transmission of livestock-associated
MRSA and showed that on two turkey farms the isolates from the animals and humans were indistinguishable or
closely related, indicating transmission.
Conclusions: MRSA carriage in individuals in daily contact with turkeys was significantly higher than that in individuals only living on the farms or than in the general Dutch population. Therefore, persons with a high degree of
contact with turkeys have an increased risk of MRSA carriage, and we propose that they should be screened prior
to hospitalization in order to decrease the risk of nosocomial transmission.
Introduction
In Europe, livestock-associated MRSA (LA-MRSA) mainly comprises isolates belonging to MLST ST398. MLST ST398 corresponds
to multiple-locus variable number tandem repeat analysis (MLVA)
complex 398. LA-MRSA can be transmitted from food-producing
animals to humans, especially to persons in close contact with
animals.1,2 In the Netherlands, the prevalence of LA-MRSA has
been investigated in healthy pigs, veal calves, horses and broilers.3 – 7 Fifty-six percent of pig herds were classified as MRSA positive, while MRSA prevalence was 88% on veal calf farms and 16%
among persons living and/or working on these farms were MRSA
positive.3,6 LA-MRSA was found on 4 of 50 conventional broiler
farms and in 8 of 145 persons (5.5%) living and/or working on
these farms. Contact with broilers was identified as a risk factor
for MRSA carriage.5 On eight organic broiler farms, however,
no MRSA was detected in humans or animals.8 In comparison
with conventional farms, organic broiler farms have more restrictions on antimicrobial use. In Germany, MRSA has been found in
turkeys, in turkey meat and in environmental samples taken inside
and outside of turkey farms.9 – 12 LA-MRSA—in particular, isolates
of clonal complexes (CCs) 398 and 9—have been identified in
healthy as well as in diseased turkeys.13,14 Information on the
prevalence of MRSA on commercial duck farms is lacking. On
duck farms, smaller quantities of antimicrobials are used than
on turkey farms and, therefore, the selective pressure for resistant
bacteria is lower. It is likely that LA-MRSA are present in commercially reared poultry other than broilers and may also be transferred to humans who are occupationally exposed to turkeys or
ducks. To date, data on the prevalence of MRSA on Dutch duck
and turkey farms are lacking. In addition, the prevalence of and
risk factors for MRSA carriage in farmers, their family members
and the employees on these farms have not yet been studied.
According to the Dutch MRSA guideline, persons who have occupational contact with pigs, veal calves or broilers have a higher risk
of being MRSA positive and are screened before hospitalization,
while persons who have contact with turkeys or ducks are not
screened.
# The Author 2015. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved.
For Permissions, please e-mail: [email protected]
58
JAC
MRSA on duck and turkey farms
The objectives of the present study were to investigate the
prevalence of MRSA on Dutch duck and turkey farms and
among farmers and their family members, to identify risk factors
and to study transmission between animals and humans.
Materials and methods
Sample collection and questionnaires
A cross-sectional MRSA-prevalence survey was conducted on 10 duck farms
and 10 turkey farms between November 2013 and September 2014.
Samples were collected by a single employee of the Animal Health
Service. Duck flocks were sampled from the age of 22 days onwards, and turkey flocks were sampled between the ages of 6 and 15 weeks. On each farm,
a single randomly chosen flock was sampled by swabbing the throats of 60
animals (pooled into 10 sample groups). In addition, samples of dust from
five locations within all poultry houses—the drinking system at the front and
the back of the poultry house, the feeding system at the front and the back of
the poultry house and the ventilation system—were taken using Sodibox
wipes (sterile cloth with Ringer’s solution; Sodibox, France). Inside each
farm residence, environmental samples were taken from an armchair, a
television remote control, inside and outside door handles and the coat of
the pet dog, if present, using Sodibox wipes (sterile cloth with Ringer’s solution). People who voluntarily participated in the study took a nose swab for
MRSA detection. Farmers completed a questionnaire about farm management, while farmers, family members and employees completed a separate
questionnaire on lifestyle and health characteristics. The study was performed according to the Dutch law on studies with animals and humans.
Written informed consent was obtained from each participant.
Microbiological analysis
Swab samples were incubated in 10 mL and wipes were incubated in
100 mL of Mueller – Hinton enrichment broth (BD, France) with 6.5%
sodium chloride for 18 h at 378C. For selective enrichment, 1 mL of broth
was transferred to 9 mL of phenol red mannitol broth with 5 mg/L ceftizoxime and 75 mg/L aztreonam (bioMérieux, France), incubated for 18 h at
378C and subsequently plated onto Columbia agar with 5% sheep blood
(Oxoid, Germany) and Brilliance MRSA 2 agar (Oxoid, Germany) and incubated for 18 h at 378C.
Suspected colonies were typed by spa typing and MLVA typing as
described previously.15,16 The MLVA for S. aureus also included markers
for the presence of the mecA and mecC genes and the lukF genes encoding
Panton–Valentine leucocidin (PVL).17
A farm was classified as MRSA positive if at least a single broiler sample
or dust sample from the poultry house tested positive for MRSA. Forty MRSA
isolates from ducks (n¼1), turkeys (n¼15), humans (n¼7) and the environments of the poultry houses (n¼12) and home residences (n¼5), representing isolates from all MRSA-positive farms, were selected and tested
for their susceptibility to tetracycline, linezolid, quinupristin/dalfopristin,
vancomycin, clindamycin, erythromycin, gentamicin, ciprofloxacin and trimethoprim/sulfamethoxazole using Etest (bioMérieux, France) and
EUCAST breakpoints. S. aureus ATCC 29213 served as a quality-control strain.
A subset of 10 isolates from humans and turkeys from the three farms
with both positive human and turkey samples was analysed by wholegenome mapping. This method creates high-resolution, ordered wholegenome restriction maps.18 High molecular weight DNA, with an average
molecule size of 250000 bp, was isolated using the ArgusTM HMW DNA isolation kit (OpGen, USA). Thereafter, DNA was applied to MapCards containing
microchannels in which DNA molecules were stretched, immobilized to a glass
surface, digested with AflII and stained with a fluorescent agent in a microfluids system. The resulting restriction fragments were sized in a wholegenome mapper and assembled into a whole-genome map in which the
restriction sites were mapped in the order in which they occur in the chromosome. BioNumerics software version 7.1 (Applied Maths, Belgium) was used
for the analysis and clustering of the whole-genome maps. The wholegenome mapping cut-off value for isolates with indistinguishable wholegenome maps was .98%. Isolates with 95% – 98% resemblance were
considered closely related. Isolates with ,95% resemblance were classified as unrelated.18
Analysis of risk factors
Questionnaires were analysed only for the turkey farms, as all humans on
duck farms were MRSA negative. The frequency of exposure to potential
risk factors was calculated for MRSA-positive and MRSA-negative individuals. Fisher’s exact test was used to test whether differences in frequencies were significant.
Results
Animals and environment
The median number of animals per farm was 18 020 (range
4100– 47000) on duck farms and 17940 (range 14 800–25 340)
on turkey farms. On the duck farms, the Cherry Valley breed and/or
Peking breed were present, while on turkey farms the B.U.T.6 breed
predominated. On duck farms, dust samples were collected from
one to three poultry houses and on turkey farms from two to six
poultry houses. The MRSA prevalence on duck farms was 10%.
MRSA was found in one throat-pool sample and in one dust sample from the poultry house on a single farm. None of the environmental samples from the duck farm residences was MRSA positive
(Table 1). MRSA was found on three turkey farms, yielding a prevalence of 30%. MRSA was detected in 17 throat-pool samples
(17%) and in 31 of the dust samples from the poultry houses
(17%). In 15 of 49 samples from the home residences on turkey
farms (31%), MRSA was detected (Table 1).
Humans
In total, 86 persons were working and/or living on the 20 farms: 43
on the duck farms and 43 on the turkey farms. Of these, 69 persons agreed to participate: 26 on duck farms and 43 on turkey
Table 1. Prevalence of MRSA on duck and turkey farms
Prevalence of MRSA by sample source, % (number of MRSA-positive samples/total number of samples investigated)
Duck farms
Turkey farms
farms
throat swab pool samples
dust samples, poultry houses
nasal swabs, humans
environment, home residences
10 (1/10)
30 (3/10)
1 (1/100)
17 (17/100)
1.3 (1/80)
17 (31/185)
0.0 (0/26)
16 (7/43)
0.0 (0/50)
31 (15/49)
59
van Duijkeren et al.
Table 2. Results from the questionnaires from the farmers and their family members on turkey farms
Variable
Visiting the poultry houses at least 2 times a day
Physical contact with turkeys at least 2 times a day
Working in healthcare
Antimicrobial therapy during the past 3 months
Hospitalization during the past year
Travelling abroad during the past year
MRSA positive, n ¼7, % (number
of participants who answered
yes/total number of participants)
MRSA negative, n ¼36, % (number
of participants who answered
yes/total number of participants)
P (Fisher’s
exact test)
85.7 (6/7)
85.7 (6/7)
0.0 (0/7)
0.0 (0/7)
14.3 (1/7)
85.7 (6/7)
16.7 (6/36)
19.4 (7/36)
5.9 (2/34)
20.0 (7/35)
8.3 (3/36)
52.8 (19/36)
0.001
0.002
1.00
0.326
0.523
0.209
farms. No MRSA-positive persons were found on the duck farms.
On 5 of 10 turkey farms, including all three farms with positive animals, one or more individuals tested MRSA positive, and on all of
these farms, samples from the home residences were also MRSA
positive. On the turkey farms, 5 of 11 farmers (45.5%) were MRSA
positive, compared with 2 of 32 family members and employees
(6.3%). This difference was statistically significant (P¼ 0.008). On
the three MRSA-positive turkey farms, 4 of 10 individuals (40%)
were MRSA carriers, compared with 3 of 33 individuals (9.1%) on
the seven farms where all samples from the turkeys and poultry
houses were MRSA negative. The prevalence of MRSA-positive and
MRSA-negative persons related to several possible risk factors is
shown in Table 2. Visiting the poultry houses and physical contact
with turkeys were identified as risk factors for MRSA carriage
(P ¼ 0.001 – 0.002), while antimicrobial therapy within the past
3 months, working in healthcare, hospitalization during the past
year and travelling abroad during the past year could not be identified as risk factors.
All MRSA isolates from animals, humans and the environment
were mecA positive and mecC negative, belonged to MLVA complex 398, were PVL negative and belonged to spa type t011. All
40 isolates were resistant to tetracycline and susceptible to linezolid, vancomycin, quinupristin/dalfopristin and trimethoprim/
sulfamethoxazole. Resistance to clindamycin and erythromycin
was common: 60% of all isolates were resistant to both antimicrobials. More than half of the isolates (52.5%) were resistant to
ciprofloxacin, and resistance to gentamicin was also common
(52.5%). Both isolates from the duck farm were resistant only to
tetracycline. The isolates from turkey farm 1 were resistant to
tetracycline, gentamicin, clindamycin and erythromycin, except
for one turkey isolate that was only resistant to tetracycline and
gentamicin. All isolates from turkey farm 6 were resistant to tetracycline and ciprofloxacin, and six isolates had additional resistances to clindamycin, erythromycin and gentamicin. On turkey
farm 10, the isolates from the turkeys and the dust in the stables
were resistant to tetracycline and ciprofloxacin only, whereas the
isolates from the farmer and the home residence were resistant to
tetracycline, clindamycin, erythromycin and gentamicin, but
susceptible to ciprofloxacin. On turkey farm 4, the isolates from
the farmer and the home residence were resistant to tetracycline,
clindamycin and erythromycin, while the isolate from the family
member was also resistant to gentamicin. On turkey farm 5,
the isolates from the farmer and the home residence were resistant to tetracycline, clindamycin, erythromycin and ciprofloxacin. Whole-genome mapping showed that the isolates
60
from the farmer and the animals on farm 10 were different
(75% homology), while the whole-genome maps of the two isolates from the animals on this farm were indistinguishable (.98%
similarity). On farm 1, whole-genome maps of the isolates from
animals and humans were highly related (.95% homology),
while indistinguishable whole-genome maps (.99% similarity)
were obtained from the isolates from humans and animals on
farm 6. Therefore, on farms 1 and 6, whole-genome maps indicate that transmission occurred between humans and animals,
while on farm 10, transmission was not confirmed (Figure 1).
Discussion
To the best of our knowledge, this is the first study on the prevalence
of MRSA on duck farms. The MRSA prevalence on duck farms seems
lower than on turkey farms: fewer farms and fewer samples were
found positive. This difference might be due to the lower consumption of antimicrobials on duck farms compared with turkey farms,
but other explanations, e.g. different management practices, are
also possible. Richter et al.11 found that 18 of 20 of the flocks investigated (90%) were MRSA positive on German turkey farms. Another
German study showed that 20% of dust samples on turkey farms,
66% of turkey carcasses and 32% of turkey meat samples were
MRSA positive.12 Friese et al.10 were able to detect MRSA in the air
outside of two turkey farms and in the soil of 44% of the turkey
and broiler farms investigated. MRSA CC398 can cause infections
in animals and has been isolated from an abscess of a turkey.13
Most isolates belonged to the livestock-associated CC398 (spa
types t011 and t034), but other spa types related to other MRSA
lineages, such as t1430 and t002, were also found. In the present
study, all isolates belonged to CC398 and spa type t011 and were
resistant to tetracycline. The MRSA prevalence among animals on
turkey farms found in the present study was lower than that
found in the study of Richter et al.,11 but higher than that found in
the study of Vossenkuhl et al.12 Possible explanations might be the
different populations investigated, different management systems
on the farms or different methods used between the studies.
Of the 43 persons associated with the turkey farms, 16% tested
MRSA positive. On MRSA-positive farms, 4 of 10 individuals (40%)
were MRSA positive, compared with 3 of 33 individuals (9.1%) on
farms where all samples from the turkeys and the poultry houses
were MRSA negative. This finding is much higher than the prevalence in the general population in the Netherlands, which is
0.2%.19 On conventional broiler farms, 5.5% of persons living
and/or working on these farms were found to be MRSA positive.5
JAC
Figure 1. Whole-genome maps of MRSA isolates (n¼10) from farmers, a family member and turkeys on three farms (farms 1, 6 and 10). This figure appears in colour in the online version
of JAC and in black and white in the print version of JAC.
MRSA on duck and turkey farms
Richter et al.11 found an MRSA prevalence of 37% among persons
on turkey farms, but, in their study, the MRSA prevalence among
the turkey farms was much higher than that in the present study.
The human MRSA carriage rate was higher among farmers on
Belgian veal farms (72%) than among those on beef farms (11%)
or broiler farms (3%).20 In the present study, farmers were more
often carriers than family members, and physical contact with
live animals and entering the poultry house were identified as risk
factors. Most MRSA-positive individuals were farmers, and both
MRSA-positive family members reported frequent contact with turkeys. Whole-genome mapping proved a good tool to investigate
LA-MRSA transmission, as all isolates in the present study had
the same MLVA type and spa type. spa typing and MLVA typing
are often used to investigate transmission of LA-MRSA, but
although this strategy is well suited for characterizing most MRSA
isolates, these methods provide very low discriminatory power for
isolates belonging to MLVA complex 398. In the present study, isolates with the same spa type and MLVA type were sometimes found
to be only distantly related using whole-genome mapping, indicating that they were not epidemiologically related. Whole-genome
mapping has much higher resolution, and, using this method, epidemiologically unrelated LA-MRSA isolates that were previously
indistinguishable by spa typing and MLVA typing can be differentiated. Other methods, including microarrays, PFGE and susceptibility testing can also be used. Transmission between animals and
humans was likely on two out of three farms. On the other
MRSA-positive farm, isolates from the animals differed from that
of the farmer: the resistance patterns and the whole-genome
maps were different. These findings are in accordance with the
results from a previous study of MRSA on dairy farms.21 The farmer
might have been colonized from a source outside the farm, or from
a different batch of animals, as not all animals present on the farm
were investigated; turkeys from previous production rounds might
also have been the source. This explanation might also account
for the MRSA-positive individuals on two farms with MRSA-negative
animals. Contact with pigs, veal calves or broilers has been previously
identified as a risk factor for MRSA carriage.2,5,22 The home environment in households with MRSA-positive persons was often contaminated with MRSA, and therefore the environment might be a source
of (re)infection for persons in the household.
In conclusion, people in contact with turkeys should be considered a risk group for nasal colonization with MRSA and should be
screened at admission to healthcare facilities in order to minimize
the possible entry of resistant bacteria in hospitals and thus
potential risks to patients and staff.
Acknowledgements
We would like to thank all farmers, their family members and their
employees for participating in this study.
Funding
This work was financially supported by the Netherlands Food and
Consumer Product Authority (NVWA), grant number V092322/01/MR.
Transparency declarations
None to declare.
61
van Duijkeren et al.
1 Voss A, Loeffen F, Bakker J et al. Methicillin-resistant Staphylococcus
aureus in pig farming. Emerg Infect Dis 2005; 11: 1965– 6.
12 Vossenkuhl B, Brandt J, Fetsch A et al. Comparison of spa types,
SCCmec types and antimicrobial resistance profiles of MRSA isolated
from turkeys at farm, slaughter and from retail meat indicates transmission along the production chain. PLoS One 2014; 9: e96308.
2 Graveland H, Duim B, van Duijkeren E et al. Livestock-associated
methicillin-resistant Staphylococcus aureus in animals and humans. Int J
Med Microbiol 2011; 301: 630–4.
13 Argudı́n MA, Cariou N, Salandre O et al. Genotyping and antimicrobial
resistance of Staphylococcus aureus isolates from diseased turkeys. Avian
Pathol 2013; 42: 572–80.
3 Broens EM, Graat EA, van der Wolf PJ et al. MRSA CC398 in the pig production chain. Prev Vet Med 2011; 98: 182–9.
14 Monecke S, Ruppelt A, Wendlandt S et al. Genotyping of Staphylococcus
aureus isolates from diseased poultry. Vet Microbiol 2013; 162: 806–12.
4 Busscher JF, van Duijkeren E, Sloet van Oldruitenborgh-Oosterbaan MM.
The prevalence of methicillin-resistant staphylococci in healthy horses in
the Netherlands. Vet Microbiol 2006; 113: 131–6.
15 Harmsen D, Claus H, Witte W et al. Typing of methicillin-resistant
Staphylococcus aureus in a university hospital setting by using novel software for spa repeat determination and database management. J Clin
Microbiol 2003; 41: 5442– 8.
References
5 Geenen PL, Graat EA, Haenen A et al. Prevalence of livestock-associated
MRSA on Dutch broiler farms and in people living and/or working on these
farms. Epidemiol Infect 2013; 141: 1099– 108.
6 Graveland H, Wagenaar JA, Heesterbeek H et al. Methicillin resistant
Staphylococcus aureus (ST398) in veal calf farming: human MRSA carriage
related with animal antimicrobial usage and farm hygiene. PLoS One
2010; 5: e10990.
16 Schouls LM, Spalburg EC, van Luit M et al. Multiple-locus variable number tandem repeat analysis of Staphylococcus aureus: comparison with
pulsed-field gel electrophoresis and spa-typing. PLoS One 2009; 4: e5082.
17 Bosch T, Pluister GN, van Luit M et al. MLVA is superior to spa typing and
sufficient to characterize MRSA for surveillance purposes. Future Microbiol
2015; 10: 1155– 62.
7 Mulders MN, Haenen APJ, Geenen PL et al. Prevalence of livestockassociated MRSA in broiler flocks and risk factors for slaughterhouse personnel in the Netherlands. Epidemiol Infect 2010; 138: 743– 55.
18 Bosch T, Verkade E, van Luit M et al. High resolution typing by whole
genome mapping enables discrimination of LA-MRSA (CC398) strains
and identification of transmission. PLoS One 2013; 8: e66493.
8 Huijbers PMC, van Hoek AHAM, Graat EAM et al. Methicillin-resistant
Staphylococcus aureus and extended-spectrum b-lactamase/AmpCproducing Escherichia coli in broilers and in people living and/or working
on organic broiler farms. Vet Microbiol 2015; 176: 120– 5.
19 den Heijer CD, van Bijnen EM, Paget WJ et al. Prevalence and resistance
of commensal Staphylococcus aureus, including methicillin-resistant
S. aureus, in nine European countries: a cross-sectional study. Lancet
Infect Dis 2013; 13: 409–15.
9 Feßler AT, Kadlec K, Hassel M et al. Characterization of methicillinresistant Staphylococcus aureus isolates from food and food products of
poultry origin in Germany. Appl Environ Microbiol 2011; 77: 7151– 7.
20 Vandendriessche S, Vanderhaeghen W, Soares FV et al. Prevalence, risk
factors and genetic diversity of methicillin-resistant Staphylococcus aureus
carried by humans and animals across livestock production sectors.
J Antimicrob Chemother 2013; 68: 1510 –6.
10 Friese A, Schulz J, Zimmermann K et al. Occurrence of livestockassociated methicillin-resistant Staphylococcus aureus in turkey and
broiler poultry houses and contamination of air and soil surfaces in their
vicinity. Appl Environ Microbiol 2013; 79: 2759 –66.
11 Richter A, Sting R, Popp C et al. Prevalence of types of methicillinresistant Staphylococcus aureus in turkey flocks and personnel attending
the animals. Epidemiol Infect 2012; 140: 2223 –32.
62
21 Feßler AT, Olde Riekerink RG, Rothkamp A et al. Characterization of
methicillin-resistant Staphylococcus aureus CC398 obtained from humans
and animals on dairy farms. Vet Microbiol 2012; 160: 77 –84.
22 Van den Broek IVF, van Cleef BAGL, Haenen A et al. Methicillin-resistant
Staphylococcus aureus in people living and working in pig farms. Epidemiol
Infect 2009; 137: 700– 8.